HI H K M l ' ‘. IHH H| 1 1 i _____—— __/_. ___’—— 1 I b l (0..) M \H #-b FURW‘QER STUDIES WETH CRUfiHED BALED HAY AND A FRELEMENARY iNVESTEGAHQN G? FACTQRS iNVOLVED EN PELLETENG HAY Thesis {'09 ‘E'E'za Dave: :3? Hi. 5. E‘sfiCE-{Efifl STHE UNEVERSETY Leon Frankiin Sanderson 1958 mum; ulzllljjflml Lu! (Ifll 111me m1 nu 1931; ll "‘. y' I. " fi’Ir- - WW FURTHER STUDIES WITH CRUSHED EALED HAY AND A PhELIMINARY INVESTIGATILN OF FACTORS INVOLVED IN PELLETING HAY By LEON FRANKLIN SANLLHSON AN ABSTRACT Submitted to the College Of Agriculture of Michigan State University of Agriculture and Applied Science in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE ,J Department of Agricultural Enqineeriné 1956 ‘ Approved 17/7 M~%é/é’ LEON FhANKLIN SANDEhSUN ABSTRACT Since hay making is subject to weather conditions, a large 1033 results due to extended drying periods and due to leaf shattering caused by the common methods of harvesting. Before this loss can be reduced a method is needed to reduce field drying time. Crushing alfalfa increases the drying rate, but the drying time exceeds the average period between rainfall in humid regions. The increase in drying rate accelerates as drying conditions improve, but the decrease in total drying time remains at from three to four hours in most cases. Tedding crushed alfalfa does not increase the rate of drying when the yield is 1.5 tons per acre. The quality of baled alfalfa when placed in an Open storage is affected by the density and moisture content of the bales and the weather conditions during the storage per- iod. Must will develop in bales With 6 pounds per cubic foot densities and 12 percent moisture content. Crushing does not affect these limits. Pelleting the hay is a possible method of solving the many problems inherent with common methods of harvesting hay, Fundamental requirements for obtaining good quality pellets are needed before a machine can be made which will produce pellets from long field cured hay, The pellet density—moisture-pressure relationship for green alfalfa is given by the equation 8 0.805 - o.ooooosess(z + 7360)(X - us.2) - a where Y pellet density - grams per cubic inch. X - moisture content - percent 2 pelleting pressure - pounds per square inch The coefficients of this equation are altered by a change in time of pressure application or quality of hay. The hay should be below 30 percent moisture content to obtain firm pellets. Pellet densities of 25 pounds per cu— bic foot are recommended for two inch diameter pellets. These can be obtained with 15 percent moisture content hay by applying a Pressure of 5000 pounds per square inch. In- creasing the weight of material in each pellet or decreasing the time of pressure application will result in a lower pele let density. Two inch diameter pellets are faster and easier drying than either chopped or baled hay. The narrow diameter re- duces the wet centers found in baled hay and also the para tial pulverization of hay stems during pelleting allows better drying characteristics than obtained with chopped haY. Firm pellets are easier to handle than baled hay and can be mechanically loaded and stored. Storage space re- quirements are reduced and pellets can be both mechanically and self-fed to livestock. Pelleting reconditioned baled hay is similiar to pel- leting green hay, The only noticeable difference is the period of time required for pressure application to obtain good quality pellets. The presence of available starches and sugars is more irportant to forming a good quality pel- let than the period of time the hay has been in storage. These nutrients are reduced after a period in storage, how- ever, and in some cases additives may be necessary to re- store the binding qualities. FURTHER STUDILS WITH ChUbHED BALED HAY AND A PhELIMINAhY INVEdTIGATIUN UF FAUPUhS INVULVED IN BLLLLTING HAY BY LLUN FnANKhIN ShNUEhSUN A THESIS Submitted to the College of Agriculture of Michigan State University of Agriculture and Applied Science in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Agricultural Engineering 1956 ACKNOWLEDGEMENTS The author wishes to express his great indebtedness for the helpful suggestions, guidance and constant encouragement of Professor H. F. McColly of the Department of Agricultural Engineering of Michigan State University. Acknowledgement is due the J. I. Case Company of Racine, Wisconsin for providing the field equipment, storage facili- ties and the funds necessary to carry out this study. Thanks is given to Professor A. W. Farrall for making the assis- tantship available. Credit is due Professor W. H. Sheldon of the.Agri- cultural Engineering Department for his interest, help and suggestions during the crushing studies. Professor D. E. Wiant made valuable suggestions on improvement of the section of the manuscript concerning the crushing studies. Professor 8. T. Dexter Judged the hay for quality and presence of mold. Mr. Andre Laurent gave helpful advice on the computation of the regression analysis for the pelleting study. Appreciation is extended to Miss Lucy Sweat for typing the manuscript. 11 TABLE OF CONTENTS Page INTnUDUCTIUN BibliOgraphy t A STUDY CONCERNING THE EFFECT OF TEDDING UPON THE DRYING RATE OF CRUSHED HAY, THE RELATION OF WEATHER CONDI- TIONS TO THE EFFECT OF CRUSHING AND THE EFFECT OF CRUSHING UPON THE SAFE STORAGE LIMITS FOR BALED HAY Reasons for the Study Review of Literature Objectives Apparatus Field Experiments Design of Experiment Outline of the Treatments Moisture Content Determination ClimatOIOgical Data First, Second, Third and Fourth Tests Analysis of the Data The Results of Tedding The Results of Crushing Storage Experiments Design of the Experiment First-Cutting Test Second-Cutting Test Analysis of the Data Maximum Conditions for HustuFree Bales Effect of Crushing on Must Development BibliOgraphy AN INVESTIGATION OF THE FUNDAMENTAL PRINCIPLES OF PEL- LETING HAY IN ORDER TO ESTABLISH OPERATING PRESSURES, MOISTURE CONTENTS‘AND THE FINAL DENSITY RELATIONSHIPS Reasons for the Study Review of Literature Objectives Apparatus Experimenal Procedure Reconditioned Hay Green Hay 111 Discussion of Results Results of Pelleting Reconditioned Hay Results of Pelleting Green Hay _Bibliography ' SUMMARY.AND CONCLUSIONS RECOMMENDATIONS FOR FUTURE RESEARCH £32 51 6s 70 73 Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure 10. 11. 12. 13. in. 15. 16. 17. 18. 19. 20. 21. iv LIST OF FIGURES Forage Crusher Used in study Forage Rake-Tedder Used in Study Weighing Moisture Samples Equipment for Determining prying Rates of Alfalfa - Drying Rates of Alfalfa - Drying Rates of Alfalfa - Drying Rates of Alfalfa - Space-Stacked Bales in an Mold Development in Bales Mold Development in Bales Pelleting Apparatus Hydraulic Press Calibration Curve Reasuring Pellet While Under Pressure 11 11 15 Weather Conditions 15 Test 1 Test 2 Test 3 Test A Open low of Crushed Ray of Uncrushed Hay Ejecting Pellet from Cylinder Reconditioned Hay Density-Moisture Relation Green Hay Density-Moisture Relation Green Hay Expansion-Moisture Relation 19 21 23 25 33 38 47 50 5o Reconditioned Hay Expansion-Moisture Relation 52 53 6o Density—Pressure Relation at 12.5 $.moisture 62 Density-Pressure Relation at 21 % moisture 62 m Figure 22. Density-Pressure Relation at 33 % moisture 64 Figure 23. Density-Moisture Relation with 3000 psi 6h Figure 2A. Density-Moisture Relation with 5000 psi 65 Figure 25. Density-Moisture Relation with 8000 psi 65 Table Table Table Table Table Table Table «F’ \N I'D P e e O\ U"! 0 VI LIST OF TABLES Moisture Contents 'Test 1 Moisture Contents — Test 2 Moisture Contents - Test 3 Moisture Contents - Test A Effect of Crushing on Drying Rate Effect of Crushing on Drying Time Effect of Tedding on Drying Rate INTRODUCTION For centuries man has dried grasses and legumes in the field in order to make these materials suitable for storage. Until recent years this practice has been satisfactory and the only improvements in hay harvesting have been these which made it possible to put up hay with fewer man hours of labor per ton. This can be seen from studying available statistics. ‘ After World War I a rural population of 31,61u 000 people produced 96,687,000 tons of hay per year (8). and in 1953 the farm pOpulation of 21,890,000 peOple produced 105,000,000 tons of bay (9). At the same time the average yield dropped from 1.58 tons per acre (0p. cit.) to l.h2 tons per acre (0p. cit.). Thus, even though this was a period of rapid farm mechanization, very little improvement in quality of hay produced or reduction of loss of nutrients has occurred. LeClerc (7) has said that 10 to 50 percent of the value of a hay crOp may be last due to nay being damaged during natural field curing with traditional harvesting methods. Dexter (3) reported that in feeding trials with hay it was found that hay quality was preserved best by rapid drying, and that an extended drying period is detrimental. 2 even though no mustiness occurs. Artificial drying reduces the loss caused by natural curing, but the increased cost is prohibitive in many cases (u). Since weather conditions are uncertain in mOst of the major hay producing regions, it is essential to the problem of obtaining better quality hay that a method for signifi- cantly reducing the field curing time be found. Even in the humid areas drying time is reduced by crushing the hay between large rolls, but in many cases this reduction is not great enough to warrent the added cost of crushing the hay (5). Along with the problem of rapid drying to procure high quality hay, are the methods of harvesting hay. Bender (1) stated that the common methods used for harvesting hay were a failure because of the damage to hay quality and loss of valuable plant nutrients. LeClerc (7) has suggested a method on which experiments have been done in England as a possible solution to this loss. Grass cakes were made from green hay with a compressing machine. Although pelleted feeds were introduced in the United States in 1929 (6), very little fundamental work has been done relative to making larger pellets or cakes. Considerable interest has been shown in harvesting hay by pelleting but much more information is needed before it will be possible to perform pelleting on the scale exhibited with other methods of harvesting (2). Hay pellets have also been suggested as a method of reducing storage space requirements (ibid). The increased cost of storage facilities has caused much interest in reducing the storage requirements by compressing the hay into dense pellets which can be fed to livestock. The study included in this thesis was preceded by eight years of research on baled hay harvesting and drying. The results of the eight years of research has indicated that further study is necessary to determine a method for de— creasing field drying time. The objectives of this study, therefore, were to investigate crushing and tedding as a method of field curing hay to obtain high quality roughage, and to determine the fundamental factors involved in pellet- ing hay as a possible solution of the many problems inherent in common methods of harvesting hay. Bibliography Bender C. B. uality Hay Defined. Aggggglggggl Engineering ourna . 28 (March 19M? , pp. 103-104. Bruhn, H. D. Pelleting Grain and Hay Mixtures. agricultural Engineering gggggél. 36 (May 1955). PP. 330‘33 . Dexter, S. T. , W. H. Sheldon, and C. F. Huffman, Better Quality Hay. Agricultural Eggiggerigg Journal. 28 (July l9u7). pp. 291-293. H0dgson,.R. E. , R. E. Davis, W. H. Hosterman, and T. E. Heinton. Principles of Making Hay. Yearbook of Agriculture 19%8 (Grass). United States ‘Dénarfmsn o r cul ure, Washington, D. C. 19M8. Hopkins, R. B. Some Effects of Chemical and Mechanical Treatments in Hay Making. Unpublished Ph. D. Thesis. Michigan State College, East Lansing, 1955. 128 PP. Lassiter C. A. , T. W. Denton, L. D. Brown, and J. W. Rust. The Nutritional Merits of Pelleting Calf Starters. Journal 2; Dair Science. 38 (November 1955), pp.l2E2-12 3. LeClerc, J. A. Losses in Making Hay and Silage Yearbook of Agriculture 1 (Food and Life United States Department 0 Agriculture ashington, D. c. 1939. pp. 992-1016. United States Department of Agriculture. Yearbook of riculture 19g2. Washington, United States Government Printing Office, 1137 Pp. . Agricultural Statisticg, 1955. Washington, Government rinting Office. 07 PP. A STUDY CONCERNING THE EFFECT OF TEDDING UPON THE DRYING RATE OF CRUSHED HAY; THE RELNTION OF WEATHER CON- DITIONS TO THE EFFECT OF CRUSHING AND THE EFFECT OF CRUSHING ON THE SAFE STOR- AGE LIMITS OF BALED HAY Reasons for the Study The quality of hay produced in humid areas such as Michigan has not improved greatly during the mechanization of hay harvesting. Much of the hay production is still lost due to damage in the field by rain or molding in storage, since an average period of two days between rainfall is less than the time required for natural field curing of hay (6). Hodgson (5) said that as much as 15 percent of a hay crop may be consumed by field fermentation losses during an extended drying period. Bleaching from the sun is another deterrent to quality hay resulting in losses of carotene. Bender (1) has said that losses as great as 50 percent of the crop value may occur during natural field curing of hay. Each of these losses is due to a long drying period and can be reduced by decreasing the field drying time. Harvest- ing at a higher percent moisture content* reduces each of * Moisture contents in this thesis are percent wet basis. these losses as well as decreasing leaf shattering of the dry hay during harvesting operations. Harvesting of high moisture content hay, however, intro- duces other more serious problems. Damp hay will respire and a rise in temperature may occur. Molds develop rapidly in warm, damp hay (8). If the temperature rises above 15$°F., microorganisms are no longer active (ibid), but spots of scorched hay will occur. If heating is allowed to proceed unchecked, spontaneous ignition may occur with the resultant loss of the hay and structure. Even if the hay does not ignite, serious losses of organ- ic matter will occur (ibid) with possible losses of 100 per— cent of pure digestible protein, M7 percent of all fats,. 9h percent of the sugars and 52 percent of hemicellulose. Hodgson (5) stated that artificial drying is successful in stOpping much of this waste, but the cost of the instal- lation and operation is prohibitive in many instances. Kleis (7) investigated the quality of hay produced in the East Lansing area. The majority of the bales inspected were at best musty, even though many of the farm operators did not realise they had any musty hay. This indicated a need for more information related to the storage of hay. Review of Literature HOpkins (6) found that crushing increased the rate of 7 drying of the leaves by the same amount it increased the rate of drying of the alfalfa stems. He stated that the only gain by crushing in Michigan would be if crushing made it possible to store the hay in less time with fewer oper- ations and thus preserve more of the orOp. However. the average drying period for crushed hay is longer than the average time between rainfall of two days (ibid) and thus crushing alone is of little value in the majority of cases. In research at Wisconsin (2) it was found that crushing decreased the field drying time during both good and average drying weather and that the decrease in drying time was somewhat greater in good weather. Heavier yields (ibid) were reported to have had smaller gains in drying rate. This would seem to be due to two factors; crushing is less effective in a heavy swath due to the cushioning effect and matting of the crushed hay becomes more prevalent. Bruhn (ibid) found that operating the rolls at higher pressures produced more effective crushing, but at the same time the tendency to clog the rolls increased.' He reported that crushing at a higher P011 speed with respect to the ground speed resulted in more effective crushing. This was due to thinning out the material between the rolls and thus reducing the cushioning effect. Multiple crushing was reported by Bruhn (ibid) to have had considerable success in Wisconsin since it increased the drying rate more than once-over crushing with no noticable increase in loss of leaves. Ramser (10) reported that crushing improved the palati- bility of hay by reducing coarse stems so that they were less harsh and brittle. Hopkins (6) made a study to determine the relation of bale density and moisture content to the quality of alfalfa hay stored in a natural draft ventilation mow. In order to obtain must free bales the hay should be dried to 20 percent moisture content before baling, and the bale density should not be greater than 6 pounds per cubic foot. He indicated that somewhat higher moisture contents and densities are allowable if a little mustiness is acceptable. Objectives This study was made to obtain essential information re— lated to improving the methods of harvesting and storing baled hay, and thus reduce the yearly financial loss which farmers sustain. The objectives of this study as outlined briefly are to find: 1. The effect of tedding upon the drying rate of crushed hay when the tedding is done immediately after crushing. 2. The effect of crushing upon the field drying rate of hay as this effect is related to weather conditions. 3. The effect of crushing upon the relationship be- tween the density of bales, moisture content of the hay when baled, and the quality of alfalfa hay stored in a natural draft ventilation mow. The field experiments were concerned With finding new methods for decreasing the time required for drying hay in the field. This consisted of experiments to determine the effect of tedding immediately after crushing upon the drying rate of the hay. Also investigated was the effect of weather conditions upon the increase in drying rate by crush- ing the hay. The storage experiments were concerned with delineating the relationship of mold development in crushed baled hay during storage to the density of bales and moisture content when baled. Apparatus Thg_Egg§gg Crusher The forage crusher used in these tests was a trailing type implement with two 6 inch x 80- inch smooth steel rolls (Figure l). The rolls were spring loaded and propelled by the tractor power take-off. This crusher did not have a mower attached and could not be con— nected to a tractor with a rear mounted mower. This made it 10 necessary to make a separate operation of cruShing after the hay had been mowed. The hay was lifted and fed into the crushing rolls byti power take-off driven, rotating cylinder with retracting spring teeth. This pick-up worked well as long as it was at the correct level above the ground surface. However it was difficult to adjust the pick-up for varying ground conditions. The crusher worked well in alfalfa and jamming of the rolls was slight. The jamming that did occur was as'a result of foreign materials such as corn stubble refusing to pass between the rolls. ‘ghe Forage Rake-Tedder. The side delivery rake-tedder used in these tests was a ground driven, cylindrical-reel type (Figure 2). In order to make it pOssible to ted a sin- gle windrow in a test plot, three teeth on the front end of each reel bar were turned up 180 degrees so that the width of raking was narrowed to seven feet. The hay swath was torn apart by the tedder and tossed into the air several feet. In observing the swaths of hay in dense areas of the field after tedding, the action of the tedder was effective in fluffing both crushed and uncrushed hay. The Bale Sampler. The bale sampler used in these tests was designed and constructed by Eggleton(h). This device consists of a-two-inch diameter stainless steel tube approx- 11 -. 1r ’"Q' " p _. heru'sed 11!“:th Stu . . Figure 1. foragew is Figure 2. Forage Rake-Tedder Used in the Study. 12 imately two feet long. A saw toothed blade is attached to the end of the tube and the device is rotated by a 5/8 inch 'standard electric drill which is mounted in a cradle. The bale is held in a rack where the rotating tube is forced against the end of the bale to cut a core sample from the bale. Field Experiments These experiments were carried out to obtain data per- tinent to reducing the required time for field drying hay. This study was set up to determine the effect of tedding immediately after crushing upon the drying rate of alfalfa. The effect of crushing upon the drying rate was also studied during different conditions of temperature and relative humidity. A twenty acre field of first year alfalfa was used for these experiments. The alfalfa had been seeded in corn the previous fall and had been cultipacked once after removal of the corn. The lack of cultivation caused the hay to be growing in narrow bands between the old corn rows, Design 2; the Experiment. This study was designed so that the effect of crushing upon the drying rate of alfalfa could be evaluated during several different weather condi- tions. In order to have a comparison, one half of the plot was crushed and the other half was left uncrushed as a 13 control. In order to determine the effect of tedding upon the drying rate, the crushed and uncrushed plots were each divided into a tedded and an untedded plot. Thus, there were four plots with different treatments. Each plot was composed of four seven-foot wide swaths and was approximately forty rods long. To reduce the effect of variations in fertility and density of growth, the four treatments of each test were conducted on one large plot. This plot was then divided into four ten-rod long blocks. The drying rates of each of the four treatments was determined by taking moisture samples from each of the blocks for each of the treatments at intervals during the drying period. During the first cutting tests, one sample was taken from each of the blocks for each of the treatments. This was changed during the second cutting tests to two sam- ples to increase the precision of the experiment. The re- sults were then analyzed as a randomized block experiment. Outline of the Treatments. 1. Mow hay, no tedding, determine drying rate, rake at #0 percent and bale at 20 percent moisture content. 2. Mow hay, ted immediately, determine drying rate, rake at ho percent and bale at 20 percent moisture content. 3. Mow and crush, determine drying rate, rake at to percent moisture content and bale at 20 percent. 1h H. Mow, crush. and ted, determine drying rate, rake at no percent, and bale at 20 percent moisture content. The treatments were raked at 40 percent moisture content to reduce the amount of leaf shattering. 391s1g2§ Content Determination. In order to obtain an accurate measure of the moisture content of the hay in the windrow a cross-section was remoVed by cutting with shears and placed in a ten-pound paper sack. The sack was then closed and labeled with an identification number. Windrow samples were weighed inside the equipment trailer within one hour after sampling (Figure 3). The samples were later taken to the laboratory and placed in‘a steam drying oven Operating at 150°F. for not less than #8 hours. The samples were weighed upon removal from the oven and the moisture content computed from this data. The moisture content of the bales was found by cutting a core from the middle of one end of each bale with the bale sampler. These samples were weighed and dryed in in- dividual sacks and the moisture content computed from this data, ClimatolOQica; Data During the first cutting tests psychrometric readings were taken at regular intervals with a sling psychrometer. Wind and cloud cover were noted and the total rainfall was recorded with a Standard Weather Bureau Rain Gauge. F igure it: ads-limes? 16 A hygrothermograph and an instrument shelter were used for the second-cutting tests in order to obtain continuous readings of weather conditions. The hygrothermograph was located about a fOOt above the ground and calibrated with an aspirator psychrometer at least twice each day in order to get an accurate measure of the relative humidity and temper- ature at the ground surface. Wind velocities were recorded with an anemometer wind totalizer located three feet above the ground (Figure h). gigs; ISEE- The first test was started on June 15, 1955. The ground was saturated with moisture after several days of heavy rainfall. The mowing was started at 10:00 a“ nu Crush-' in; and tedding were done as soon as mowing was completed at 11:30 a. m. The hay was sampled at intervals for moisture content (Table l) and was raked into windrows at 1:00 p. m. on June 16. The treatments were all baled on June 17 at 3:00 p. m. gecond Test. This test was mowed at 9:00 a. m. and and crushing was started at 9:“5 a. m. , June 22. The tedded swaths were completed by 11:00 a. m. and moisture samples iwere taken at intervals (Table 2). The hay was raked at 3:00 p. m. June 23 and baling was started at 1:30 p. m. the fOllowing day. June 2h. .22229.22§2i The second-cutting tests were begun on July 20. The hay was mowed, crushed, and tedded at 10:00 a, nu: 11:00 a. m.; and 11:30 a. m. respectively. Weather 17 and moisture content data were taken as before, but two sam- ples were taken from each block for each treatment (Table 3) to increase the precision of the tests. The hay was raked at 3:30 p. m. the same day and baling was started at 10:00 a. m. July 22. Fourthslg_g. Mowing, crushing and tedding were started at 9:30, 10:30, and 10:u5 a. m., respectively, on July 2H. The drying rate was determined by taking moisture samples at intervals (Table #2 and the hay was raked at #:30 p. m. the same afternoon. Rain had been forecast and the baling was delayed in order to determine the effect of crushing upon the drying rate after a rain. No rain occurred, however, before the hay was baled at 11:00 a. m. August 1. Analisis g: the page, A complete analysis of variance was carried out on all of the data reported in this experi- ment. Each set of moisture content samples was analysed as a randomized block to find significant differences between moisture contents of hay from various treatments. The average moisture contents were then plotted to show the dry- I ing rate of each treatment we h moisture content versus drying time for each test (Figures 5, 6, 7, and a). The drying rate during the first day of eaCh test be- fore it was raked was computed for each treatment (Table 7) These rates in percent moisture loss per hour were then an- alyzed using the F test with treatments versus days to 18 Table‘ 1. First butting moisture Contents - Test 1 June 15, 1955_ Time Uncrushed I Crushed i 3' 1 Day Hour Untedded Tedded fUntedded ; Tedded ! I II a III 1 IV (June AVerage moisture content ~percent 1 15 12:00 AM 70.93 69.6“ 67.09 ! 68.99 E 2:30 PM 57.78 60.22 54,1h i 55.82 “* 4:00 PM 5n.14 53.34 50.31 { us.35 i 16 10:00 AM #6.42 #2.85 40.41 44.02 [a x 1:00 PM 39.32 no.71 28.78 30.67 i** f 3315 PM 32.73 33.89 30.50 31.ns 4:30 PM 31.52 34.51 24.01 2k.91 ** 17 9:00 AM 2s.c4 30.91 25.47 22.19 10:00 AM 3 23.41 | 25.74 18.60 I 17.0% I 2:15 AM { 17.h2 ( 22.42 1a.78 17.21 i 1 A11 data average of four samples except 10:00 AM June 17 average of two samples. lyzed as a randomized block. Each set of moisture data was anau Each set of moisture content data which indicated a significant difference in moisture content as a result of crushing is starred. * Data significant ** Data highly significant ..... -1 111 11. -111--- I|[(l 1.00 IYO(. I . | - ----.- - _ H _V——-..--1-1__.-,_.__ --.-.whw , u . v : . , . ...-.. .H-1._._.-- 1-9--..- _. 1 -....J :n .(IIJI.(!¢ Q _ . Iii?! 1-- e . . _ .Ycl .+ -1951. .. , . _. .. ... . 1.1— ... ._¢.__..._ .. ..---.» . - __1H—- --.... ..- -<>--~- -.- -._1+-_-_....- - 1 1-_> - --— - .... 1 p . 1 --._,.-..1 ..-_... f 1 11.1 111.... .1 i 1 - .1 __,1__.1-.1111._1_-,.-1_ - ! ...rd ..T 1 1 r 1 _ _ 1 1 _ 1 u- a! I. 1 1 1 a .1 1 _ v 1 1 1 . : . - . | . D -L._1_1_~._. .____.,._._.__.._h,. - . 1. ». .... , . I . ' — . —- q . .i ' I "f. ‘I . . I I ~. . - 1i . . . . 1 . -. . . _ 1.. 1 ..‘ . .1. . //\ -- -; . » ~- I - I ‘ - l v 1 . i {. l. v i "1.11111111- .- ' r I I .‘ , I , .' _. -.__ ¢_._~» » p -..VnJ—efl ___.__.~..-..._._. e—— -...._4_‘..._- -13. ... ___.....+_..- ‘ ...—..- .... 1 ‘ - . ‘ n : a . . . _ . . I l f 4 I 1 i I z r u 1 .1 ~. e. 1 h — » _ I?! V . a} 1 .. e ' ' I . -1. 11-- .9“... .11-..-- . . .|.Irtl.v2.D 1.11": . vi: . $11.11.-- 11 -1 111-13 0‘o0000se 13‘! .trvli’lcl I\ + 1... ”"7" .1 14. i J L. :72: 11-1.1.4 ..-. - . ...... IAIII"+< I: L £—-»-.-. ... I l I V v ........ ...... . Rim :m/E . awn/7V I“; I ..xiirti vnmflwava‘ 1‘.- / . .--1-...-.. I I . I 723:7: 1 l 1 L Jr , "w; 4-. F/Rdrr' can-m ".4 ..Ag : (ITO/V5 ,,. 1 I 1* Y 3.4 res-0524234112: a..- I ( .111...- :r-e~~~—*a~—« BED 3.4 ‘9‘" ---7--—' 1* 1.._._1‘ 1.4!..-111111— , :g ... mm ......z 1 mt 1 i l v ! 4 1.41 l”! 4 1». 7b..” ..~L.1- . 11141-14 . 4... 4 .4 ......4 ...4w-114 - . .4. ..- . - .4.. . . 4.... _ 4. 14... . . 4 . .. . 4.. 4- . 4 . . . . .- 1 1.4-41.1- .q 4. 1 141-12.41-11--- 1- 4 ..... ..11 -1. .4- .4 1 - 141114.11-.. - 1 .0141 4... 4 . 4 . 4 . 4 4 4 ...4 4 _ A 114 4 -11. .- 41. 4 4 4 . 4.- I .10 r .-.; - . 4 4 .1 .4 A I. . 4 .. 4 s -_1 4.. 1 .1 ...... :4 4. 4 .-.--14.-- 1V1-1L1. .. 4 4 4 4 4 ./4 A . 4 .4 4 4 . .4 2 f F 4 4 4 4 .. x. 17-. I. 1.1.... 4 ... 4 4. 4 ..., 1A1- 4 4 . 4 . _ 4 111-1111111111 11111 11. 1 I 1111 1111:: ,111 111 1111 411, 11 1.. 4 V n 4 4. .4 4 4 / f 4 _ 4 4 . 0 M 4 .9 4 44a 4v .4.:- 1 r 111.1 |111 .11 41111.1 _ . A 4 4 .H h a 411 . _. 4. 4k\ .4 0 . . .. 4 4 4 4 . 4 ... / . .5 ::.;-_ ...... 4. ..._. - 4 4 .... 4 -1--1...; - _ 4 . 4 . 4 ._ T 1111. . 1 1. 1...- - I a .- 1 1 -.1111--1-1-111l.--4 111. 1- 11 - 1 1 . 4111-1111- 4 DJ . 4 .4.. _ 4 _ .4 1 . _ 4 .4 . 4 4 . 1. Rm 4 4 .4. 4 4 44 4 4 . 4 . 1. 1 .411 I 4 4 p 4 -. ...-..111 11111 ...-...11 . 4 4 4 4 . 4 4 4 . 4 4 4 4 4 4 4 4. . 4 4 .1 4 1 4. 1 1. . 1.4 4L raw 4 6 4 . .. .. . . 4 H .. . 4 4 4 1 -44- 44 . .H .. 4. .u 4..*.4--41N1.- 4 H 4 .J _ H . . .. 4 4 4 4 a . _ . . / . .. . . 4 .4 4 1111 141111111 -11. 1 1. -1 11- - -4 :1 1.- 1 1-111 - 1- ---..11-1 .1 1 1 1-1.1.711 - ... .1 1 4114-1;- 511144-V411 . 4. _ .. 1 _ 4 4 .4 H 4 . .1 - -. .4 - .4. R 4 . . . .H 4 4 4 4 .4 .4 . 4. . 4 0 4. . . 41 .. .(r . 1 111 . 1 w 4 11.1 AN 1 .4 ‘1 1 4 4 4 . . 4 4. 4 4 1 m, 4w 8 4 4 4 . E. D . 4 . 1- 11. 11.... -. 1 - 4 11 - 111 -1.”- 1 Nun-11.11101 {al-11111flw1. - -1- 4111-. 4 4 4. - 7 .0: 4m 9 4 6 .. 4 4 ../ E 2 . AI. . 4 4 . . 4 4 W E . .5 E Y 4 4 . -R-DD4H1H. - ..L-41 4 4 .H U E E .. U 4. 1111 - .. 4 - ..--TM1M-5-$1. 11.1-1.1111 4 E A U U . 4 . a 4 wnoosfl 1:4. ..... 4 -4 T E M C ..C 91. 14. .4 4 AF 4 / 4 4 L M - a 4 a - 1 4. 11 111.-- .1-. -.-1- ..1-111111 4 4 . 4 - . 4 - 4 - 4 ._ . 4 H 4 / 4 _ - 0 4 0 4 0 4 0 .. . 0 4 0 4 4 4 9 4 8 . . . 4 4 .4 . .. “a :2 _- .4 .7 . Hi :4 .- . - .4. .4 .. 4 . . 4 4 . 4 4. 4 _ 4 4 . 4 fimrzmxtflxo 4 . nah-G Q m1 1 4 , 111 -‘ 4. T1111 161111-1111.- 1111111J11 1111.1. 1-11 11111-11411-1-1 1.11 11411 . 111.. 111.7111. W111: 1111. - 1 ..1111 1 - - - v .1 11. 1'1. .- ...-1111 1|. 1. 111 1.1 1 . - - 1111111. -. 12- 1 1111 1. 1 1. . 4 .44 \QESQ. 4.0.514 Nut-4K4 133.4%. km; 4K>44w1§€0 numb-REE 4 4 4 , 4 4 4 mkcfiwwmaxfifi $4.4 4 4 4 4 4 .4 . 4. . . 4 .4 1. ... _- 1..-..471 / 1 a. T 7. .hU-L ZwmuNL-i434 mzmu4141w QU1¢.Q IQx‘QD T4UZ4 gun. ON VA ON ZWmUNFU.D ON 1 1.14ovm .DZ 26 Table 5. The Effect of Crushing on Drying Rate as Related to Weather Conditions Weather D 1 But Date 7‘ ry ngfi' e Conditions Uncrushed Crushed Increase of Temp. Humidity 1 I III III over I High 10' 2 .--. —+~~ ‘ ‘ Percent moisture loss per hour“ 9? Percent {June 15 “.197 1.340 0.1h3 78 #8 'June 22 2.460 2.366 0.h06 73 16.5 July 20 1.920 8.000 3.030 89 10 :July 29 6.115 7.167 1.352 90 34 .Average 1.123 5.667 1.213 * Drying rates for the first day before the hay was windrowed. Table 6. The Effect of Crushing Alfalfa Upon the Drying Time 1 Weather Total Drying Time Conditions Test Uncrushed Crushed Decrease of Temp. Humidity I III III from I High Low Hours °F Percent l 50.00 ”7.75 2.25 78 #8 2 55.00 52.00 3.00 73 16.5 3 28.25 21.50 3.75 89 no 29.25 26.25 3.00 90 311 27 oflukofi Iced; was man map waommn madzhu we had pagan map now mopua madman t m . . wwwnm>4 _ and 0 wow m moo.m oaa.o :mm.: :m:.: Haapm>o . . M wzdppso cam mma o- mom.~ :m~.~ ” ma~.o 5mm.o mam.m no owmnm>< _ mam.o mmw.~ Nm:.~ M mmm.o oa:.m maa.m mm hasa _ :mm.ou m::.~ oo.m “ m:~.a moo.m omm.: om hash . unappso panda .mwmhm. emo.: mow.m * wm:.ou Hmw.m mmm.m no mwdpmpd wma.o :mo.m mow.m :ww.ou mwm.a om:.m mm mesa mmm.0 mmo.m o:M.: m mma.ou mpo.: ~ma.: ma mash :Lson pmq mmOH manpmdoa pcooamm HHH nmpo >H >H HHH H ampo HH an H omdopocH cocuoa uwdoopca omGogoCH caucus umduopaa umsmsuo cmnmsuo poz muan mam accumpcs mzmno> caucus Ho mopdm wadmun uuauua<_uo opam wcamua ao.wc«uuoa no poouua «ma .N manna determine if any of the treatments had a significant effect upon the drying rate of alfalfa. The t test was then used to determine whether tedding or crushing was effective. The average drying rates and the differences between the drying rates of treatments for the four tests were then cOmpared (Tables 5, 6 and 7). The Besultg g: Teddigg. At no time during the tests was the effect of tedding upon the drying rate of crushed hay significant when the tedding was performed immediately after crushing. Although decreases in the drying rate as well as increases occurred, none of these differences were large enough to be significant (Table 7). when the drying rates of tedded and untedded treatments were each averaged for the four tests, tedding caused an in- crease in drying rate for both crushed and uncrushed hay. Tedding increased the drying rate of uncrushed hay 0.110 percent moisture lOss per hour. The increase in drying rate of crushed hay as a result of tedding was 0.139 percent moisture loss per hour. This difference in increase in dry- ing rate due to tedding between crushed and uncrushed hay was negligible. It was observed that in areas of the field where the alfalfa yield was light, very little matting of the crushed alfalfa occurred and it was difficult to determine whether a swath had been crushed without inspecting individual stems. 29 Where the hay was denser, matting was frequent and the crushed swaths were easily distinguished from the uncrushed swaths. In these dense areas the effect of tedding was also much easier to observe than in the light areas since the swath was in a more ruffled condition after tedding. The hay yield per acre was below the average for Michigan of 1.5 tons per acre (9). The first cutting averaged 1.45 tons per acre, while the second cutting produced only 1.05 tons per acre. If the hay in the test plot had been denser, tedding would have had a greater effect, but this effect would not be significant unless a much larger sample of moisture content was taken. ‘Ihg Results 2; Crushing. Crushing caused an increase in ' the drying rate of alfalfa in each of the four tests. The, four tests had an average increase in the dryng rate of 23.3 percent as a result of crushing on the first day of drying before the hay was windrowed (Table 5). The analysis of variance indicated this increase was significant. It can be seen (Table 5) that as drying conditions im- proved, the increase in drying rate due to crushing became greater. During poor drying weather with drying rates of H.197 and 2.460 percent moisture loss per hour for uncrushed nay, crushing increased the drying rate only 0.1u3 and 0.906 percent moisture loss per hour, respectively; whereas during more favorable drying weather with drying rates of 3o n.920 and 6.115 percent moisture loss per hour for uncruShed hay, crushing increased the drying rate 3.080 and 1.352 per- cent moisture loss per hour, respectively. The fact that the test with the lowest drying rate did not indicate the lowest increase as a result of crushing is due to two main factors. The most important is that the effectiveness of the crusher depends on the amount of material passing through the rolls. The amount of material varies constantly, therefore the effect of the crusher varies also. The other factor affecting the results is the difficulty of getting an accurate measure of the moisture content of the hay. More than twice as many samples would have to be taken as were taken during the second cutting to remove the large error due to variation in moisture content of the hay in the plot. Another method of observing the effect of crushing as related to weather conditions is to compare the hours of drying time required to reach twenty percent moisture content and the decrease in drying time as a result of crushing. It can be seen (Table 6) that although the tOtal hours of dry- ing time increased greatly for poorer weather conditions, the decrease in drying time as a result of crushing did not vary appreciably, except in one test. The decrease in drying time was not less than two or greater than four hours as a result of crushing except in the 31 third test. Crushing indicated a decrease 0f eight hours in drying time in the third test. Storage Experiments This study was made to determine the effect of crushing upon the relationship between density of bales, moisture content of hay when baled and the quality of hay when stored in a natural draft ventilation mow. These studies were made in the Case Hay Laboratory at East Lansing. The hay for the tests was obtained from the same field used for the field experiments. Design 2; the Eageriment. This study was made with crushed and uncrushed hay in order to evaluate the effect of crushing on the deve10pment of mold in stored baled hay. Both crushed and uncrushed hay were baled over a range of moisture contents. The moisture limits were set by previous work (6) which indicated that hay would be moldy when baled above 35 percent and would be must free when baled less than 20 percent moisture content. In order to get a range of densities, the tension on the baler was varied from loose to tight at each of the four moisture ranges chosen between 35 and 20 percent moisture content. Briefly, the procedure was to bale four windrows of crushed hay and four of uncrushed hay at a desired moisture range. The first windrow was baled with the tension springs 32 on the bale chamber completely loosened; the second windrow was baled with the springs Just snug; the third had three to four additional turns applied and the fourth had the maximum pressure allowed without breaking the baling twine. Each bale 'was tagged to identify it. The bales were all taken to the hay laboratory where each bale was weighed, and a core sam- ple removed for moisture content determination. The bales were space-stacked in the hay laboratory (Figure 9). This was done by leaving spaces up to six inches wide between the stacked bales which allowed the natural circulation of air around each bale in the stack. This is called natural draft ventilation, or natural mow drying. The bales were inspected for must after a period of 12 to 16 weeks in storage. A scale which classifies the extent of must development was set up for grading the hay. These grades were: Must free — no must present Slightly musty - a trace of must present Muuty - bale has decidedly musty characteristics or is moldy. tipsy-gutting Test. This experiment was started on June 27. The hay was mowed at 2:00 p. m. and crushed immed- iately. Both plots were raked at 9:00 a. m. on June 28. The baling was started at 2:00 p. m. June 28, and was com- pleted at 2:00 p. m. June 29. 33 1. (m. l Win mm. "I." Figure 9. Space-stacked Bales in an open Mow. 31+ Core sampling of bales and weighing was completed by the evening of June 30. These bales were stored in two closed bins of the hay laboratory. The dimension of one mow was 12 x 20 feet and the other was 21+ 3: 20 feet. The walls and floors were covered by plywood and the only circulation of air was through the open doors and the roof ventilators. The average daily temperature high during the first Week of storage was 72°F. Precipitation occurred on four days of the first week with a total rainfall of 0.66 inches for the week. The United States Weather Bureau twelve-year average for East Lansing was 71°F. and 62 percent relative humidity for the month of July. The bales in this test were opened for inspection after 16 weeks of storage. second-Cutting Test, Hay was mowed at 10:00 a. m. and raked at #:00 p. :11. July 25. Baling was begun the following day and was completed by 12:00 a. m. July 27. The humidity had been very high during the latter part or the test and 0.15 inches of rain fell immediately after the hay was sheltered. The sampling and weighing was com- Pl eted by the evening of July 26. These bales were stored in an open bin of the hay laboratory. Two sides of the bin were Open and the floor consisted of a one inch wire mesh 1ayed over the Joists. The average daily temperature high during the first week or storage was 82°F. and the average daily relative humidity low was 50 percent. The united States Weather Bureau Twelve- year average for August was 69°F. and 65 percent relative humid- ity. The bales in this test were opened for inspection after twelve weeks in storage. ' Analysis 23 2235. The results of the storage tests were compiled and plotted graphically with moisture content at time of balina: Versus bale density. The quality of the hay in each bale was indicated by a symbol. The upper and lower limits of the three grades of must development for crushed (Figure 10) and uncrushed ( Figure 11) alfalfa were then drawn. Maximum Moisture Content and Bale Density for Must Free Bales. In preliminary tests during 1952 Eggleton (it) found that must and mold free baled hay occurred at densities be- low 7 pounds per cubic foot. Hopkins (6) continued the tests in 1953 and reported that during storage periods of eight Weeks must occurred at not lower than 19 percent moisture °°ntent. The results of the storage tests during 1955 indi- cate that must occurred in hay which was stored at 12 percent moisture content during a storage period of 16 weeks. Wright (11) reported that must occurred in hay at moisture contents as low as 13 percent. He stated that it may take as long as 300 days to appear. This indicates that as the storage period is extended must will appear in bales with lower original moisture 36 contents. This is understandable since must may develop on hay that is dry enough to prevent mold. if the humidity of the air surrounding the hay rises above equilibrium With the hay (3). Dexter (ibid) stated that equilibrium for 1“ percent hay would be 75 percent relative humidity. Over longer periods of storage more opportunity will occur for humid conditions above equilibrium. 13.21.1119; 9! __a___usCru m 9.229 £25.15 D__ev_e_210 ment. Crushing did not indicate a significant lowering of must development. Must-free bales did occur at slightly higher moisture contents and densities for crushed hay than for uncrushed hay. However, since this difference is only one percent, crushing is of less importance in reducing molding after the hay is stored than other factors such as temperature, relative humidity, density of bales and moisture content of the hay. Differences Between 9223 and Closed EEEE- There was no conclusive difference between the limits for must development in the bales stored in the open mow When compared With the bales stored in the closed mow. This indicates that drying conditions within the space stacked bales were not limited by the circulation within the bin as much as by the density of the bales. . Accuracy 2; Limits 2; gust DevelOQment. Although each limit indicated was decided by results from not more than two test bales, a linear relationship is shown between den- sity and moisture content for the development of must in \ F" R \ 4‘0 \ \ B LBS. PER CgB/C‘ F007 N ‘0 4 (h BAA E DEALS/T Y -. N 6‘ BALE QUALITY x - MUSTY 0 -- SLIGHTLY MUSTY - I o - MUST FEEL: 12 I4 /6 I8 20 22 24 2.6 26 30 32 34 PERCENT MOISTURE CONTENT FIG. /0 THE RELATION BETWEEN BALE DENSITY, MOISTURE CONTENT AND KEEPING QUALITY or CRUSHED ALFALFA /§_ T I " a a" K 3/3F_ _s 3 X 6 $12 Q " x m“ H‘___ x q % x _T Q x ' IO I X > ‘: :3 g 9;.0 05 N— E “9 8I VIA w l. / 4 xx ,JV3 1* V , on 7__ x 9 __ 6 shay, BALE QUALITY 5 //”// X - USTY :,// e - LIGHTLY M057"? ’ O — UST FREE 4 l L L I , l /6 I8 20 22 24 as as .30 .32 .94 .36 is PERCENT MOISTURE CONTEN7 F/G. II THE RELATION BETWEEN BALE DENSITY, MOISTURE CONTENT AND KEEP/N6” QUALITY OF UNCRUSHED ALFALFA I355 153 39 baled hay. The maxinum difference in slope of the limits for must development during the 1955 tests is less than 6 degrees. This indicates that the limits are quite accurate in lepe as well as being linear, but there is no method to sub- stantiate this indication. The upper limit 0f slightly musty bales in Figure 10 was drawn parallel to the upper limit of must-free bales on the same figure since the other four limits are nearly parallel. The limits indicated {Figures 10 and 11) are only those of this sample, however, and a larger sample would not neces- sarily have parallel limits nor limits at the same mOisture contents and densities. 10. 11. no BibliOgraphy Bender, C. B. Quality Hay Defined. égricultural Eggi» neerigg Journal. 28 (March.19fl7), PP. 103-10h. Bruhn, H. D. Pelleting Grain and Hay mixtures. Agricul- tural Engineering Journal. 36 (May 1956), pp. 3353331. Dexter, S. T. The Vapor Pressure or Relative Humidity Approach to Moisture-Testing for Safe Farm Storage of 0 Harvested CrOps. égronomy ournal. #7 (June 1955 pp. 267-270. Eggleton, C. H. The Use of, and Equipment for Applying Mold Inhibitors in Baled Hay. Unpublished I. 8. Thesis, Michigan State College, East Lansing. 1953. 85 no. leaves. Hodgson, R. E. R. E. Davis, W. H. Hosterman, and T. E. Heinton. Principles of Making Hay. Yearbook of Agri— a , United States Department of Agriculture, Washington, D. C. 1948. PP. 161-167. Hopkins, R. B. Some Effect of Chemical and Mechanical Treatments in Hay flaking. Unpublished Ph. D. Thesis, Michigan State College, East Lansing. 1955. 128 p Kleis, R. W. A Survey to Determine the Quality of Baled Hay Produced by Local Farmers. Unpublished Supple- ment to the 19H9 J. I Case Company - unchigan State College Hay Curing Report. 1949. 6 no. leaves. LeClerc, J. A. Losses in Making Hay and Silage. Yearbook of agriculture 1939 (Food and Life). United States epar ment 0 ricul ure, as ington, D. C. 1939. PP. 992-1016. Michigan Department of Agriculture. Michigan Agriculture a1 Statistics, 195”. East Lansing. Ramser, J. H. and R. W. Klein. Hay Crushing for Faster Field Curing. Illinois Agricultural Extension Circular 693. 1952. Wright, N. C. The Storage of Artificially Dried Grass. Journal of Agricultural Science. 31 (19h1). pp. 19s. 211. #1 A PRELIMINhRI INVESTIGATION OF FACTORS INVOLVED IN PELLETING HAY Reasons for the Study Even though much research has been done to solve the problems related to the present methods of harvesting hay, many unfavorable factors still exiflt. Baled hay is difficult to handle in the field and in storage. Hay that otherWisc is safe for storage will mold in storage after it is baled (9). Artificial drying of baled hay is difficult and expensive. Feeding baled hay is tedious and it can neither be mechanically fed nor self-fed to live- stock. A large percentage of the leaves are lost during baling and even more are lost when the bales are torn apart for feeding. Chopped hay, however, is easily handled by machinery in the field and can be both mechanically and self-fed to livestock. A large loss of hay results from cattle refusing to eat the large amount of chaffy feed resulting from chOpping and a further loss occurs from cattle trampling the hay due to the loose quality of the feed. In many cases chopping is not acceptable to the farm operators due to the extremely dusty character of the feed. he In recent years there has been a great deal of research carried out to determine the value of pelleted feeds and LeClerc (12) has suggested pellcting hay as a possible method of solving the many problems present with other methods of harvesting. 'The research to date on pelleted feeds has indicated several advantages to feeding pellets. Eaton (6) found that calves consumed larger quantities of feed in the form of dehydrated pellets than in the form of long field cured hay, and that the calves grew faster as a result of this increas- ed consumption. Blosser (2) reported that cows do not go off feed as often when eating pelleted hay as when eating chOpped or ground hay. This allows heavier feeding of livestock. Bruhn (3) stated that pellets are easier to handle than long loose or baled hay and that hay when pelleted occupies less storage space than when handled by other methods. Most of the research on pelleted feeds has been done with pellets less than 1/2 inch in diameter. The cost of grinding hay and pelleting into these small dimensions has been questioned when feeding hay to large ruminants. Theron is considerable interest in a machine which can produce lar- ger pellets from hay in the windrow with a capacity comparable with forage balers and chOppers in use. No machine is available, however, which can produce pellets over one inch in diameter from unchopped hay. 43 heview of Literature In processing ground feeds into pellets, Grahek (8) found that there are certain fundamental requirements of any formula to obtain good quality pellets. One of the most mportant of these is the binding quality of the ingredients, such as the presence of starches. These break down into sugars during the pelleting operation and form a very effective binding agent. Bruhn (3) reported that green nay can be pressed into firm pellets as large as two incnes in diameter without the addition of any binding material or any prior treatment. With ground feeds Grahek (op. cit.) found that making pellets Without a b'nding material was less efficient because of the excess pressures and power required. He indicated that conventional pellet mills were now operating with pressures from 4000 to 10,000 pounds per square inch. From the results of Bruhn's study (op. cit.) it is indicated that additional binding material is not necessary when pelleting hay. He obtained pellet densities of ho pounds per cubic foot with pressures of £000 pounds per square inch. When the pressure was greater than this pellets were not easily eaten by livestock. if pellet diameter is reduced, trans; (op. cit.) found that a pellet mill 0f given die speed and harsepower Will still maintaining pellet quality output wril (1'; U) 'n 3*“. «TA lulu- EMS an or hardness. This result was with ground formulas, however, and may not be the same for long hay. These results indicate that much more work is necessary relative to the fundamental pressure-moisture relationship and the final density of the pellet as well as the time of application of the pressure before a portable pelleting machine can be perfected. Objectives The study was made to determine the requirements for obtaining_firm, palatable pellets suitable for feeding to livestock. Briefly, these objectives are to determine: 1. The effect of pelleting pressure upon the expansion and final density of two inch diameter hay pellets. 2. The effect of moisture content of hay at time of pelleting upon the expansion and final density of two inch diameter hay pellets. This study was made to obtain fundamental information related to pelleting hay as a preliminary study of an over-‘ all project for the possible development of a pelleting machine. 45 Apparatus Reconditionigg_9hamber. The reconditioning chamber was constructed so that a high humidity could be controlled around the hay. A barrel was used with a steam inlet and outlet..A false bottom was installed to keep the hay above any condensate and the cover was connected with a rubber casket to insure a moisture tight seal. Pelletigg Apparatus. In order to determine fundamental principles of pelletint it was decided to use a hydraulic press to apply pressure to a hay sample in a cylinder and ram device (Figure 12). This allowed accurate measurements to be made and results could be reproduced and checked with very little error., The load exerted by the press had been calibrated with a pressure gauge certified by the United States Bureau of Standards and these results were extrapolated so that the load exerted by the press could be found by merely measuring oil pressure. The pressure gauge used for the tests was cal- ibrated and these results were combined with the press calibration (Figure 13) to make the measurement of load on the ram. The pressure on the pellet was calculated from the area or the cylinder and the load on the ram. The pelleting cylinder was constructed with a standard two-inch pipe. This was placed on a flat horizontal base and the ram forced down into it. With hay above 30 percent Figure 12, Pclleting Apparatus: Cylinder, Sleeve and Base, and Ram. In HYDMUZ/C FRI? 55 CAL/BHAT/fl/V CUR VE LOAD 0N PELLET RIPS A: ON Om. WMQZfixxnc uxubeQxI \S qucmwutd \HQ Ox MQDQPK QVMVQEN 09$ 00% 3.3 034 8mm 93‘ com own 0&0 cum. Ofis Q A _ __#__l____a J .IL .IL 1 l 1 0 cube oootN ooosv snow coco soon snow cogs asses soo<\ coowd ssow\ scans IS'J 13773d NO BUNSS‘BUd #8 moisture content this device was not satisfactory. The damp hey acted as a hydraulic material when pressure was applied and oozed from the chamber between the base and the bottom of the cylinder. To correct this a sleeve (Figure 12) was welded to the base. A loose fit was made between the sleeve and the cy- linder so that when pressure was applied the cylinder would not lift from the base. This allowed much higher pressures ,to be obtained with moisture contents above 30 percent with- out loss of material from the cylinder. Experimental Procedure Pelletigg Reconditioned Baled Hay. These tests were made with reconditioned baled hay before green hay was available frOm the field. This was done to determine oer. tain fundamental characteristics of pellets early in the studies and aid in the develOpment of the apparatus. In order to experiment with hay over a large range of moisture contents the hay was reconditioned by holding it in a high humidity atmosphere until the hay was at the desired moisture content. This procedure was not satisfactory, how- 'ever, since the hay did not absorb moisture uniformily. Part of the hay was quite dry while other areas were soaked. The hay acquired a dark brown color similiar to that subjected to severe weather damage. This indicated that valuable dry 49 matter and nutrients were lost in the reconditioning process. When the hay had reached the desired moisture content in the reconditioning chamber. a hay sample of desired size was inserted in the pelleting cylinder. The pelleting was done by using a hydraulic press to apply pressure to the ram on the hay sample. ’The maximum pressure and the period of time held on the pellet was then recorded for each pellet. The length of the pellet while under maximum pressure was found by measuring the protruding length of the pelleting ram (Figure 14). The pellet was then eJected from the oy- linder (Fiure l5) and weighed. After the wet weight was recorded the pellet was placed in the electric drying oven, operating at 150°F. The pellet was reweighed at intervals until it had reached a constant weight. The first and final weights were used to determine the percent moisture content of the pellet at the time of pelleting. The percent linear expansion of the pellet was computed from the minimum and final length measurements. P eti Green Ha . As soon as green hay was avail- able the studies on pelleting freshly cut hay were begun. In- order to pellet hay over a range of moisture contents. the freshly cut hay was placed in an electric drying oven until it had dried to the desired moisture content. The sample was then inserted in the pelleting cylinder and the desired pressure applied. The time of application , .c.1 e . . . , , ,, fir“ . S~ 7 Figure l . Measuring pellet length under maximum pressure 1.“- ".- Figure 15. Electing the pellet from the cylinder 51 of the pressure and the minimum height of the pellet was re- corded. The pellet was then ejected from the cylinder and weighed. The pellet was then returned to the oven until completely dry in order to determine the moisture content when pelleted. The pellets were removed frOm the oven when dry and the final weight recorded. The pellet was then measured for both length.and diameter to be used in comput- ing expansion, final density and final volume, Discussion of Results Results of Pellgtigg Reconditioned Hay. The linear 01h pension of each pellet was computed from the length of the pellet when maximum pressure was applied and the length of the pellet after it had been dried in the oven. The density of the pellet was determined from the final volume and the percent moisture content was computed from the wet and dry weights of the pellet. This data was plotted to show the expansion-moisture relationship (Figure 16) for reconditioned baled hay and the density-moisture relationship (Figure 17). Because of the small number of samples and the large variation between pel- lets. no regression lines were computed for this data. These tests indicated that the final density of the pellet is determined by the density when pressure is applied 52 L/NEAR EXHflNSIO/V-MO/STURE REL/1 T/ON RE C 0ND! Tl ONE 0 HAY \ 3000 PSI 6000 PSI 0000 P5] 4 D ... \ 5' 00 O c 8 O c 0 § "3 a: 3 NOIS'IV var: scram/v .uvyaarsu 35' 25' 45' 40 J0 20 [5 PH? uwr MOISTURE co~rr~r FIGURE [6. 53 [5’ O s RECOND/T/ONED . HAY t I l I ”a l u” 0 13/ ‘ R“ “3% 4 Q90 000 h QQQ 0: 'DUQ) * 1 0+ 3 / + ——-:2 +- on n / a J l I J \o In V? '9 N 'r 0 Q N ”as” exams are! Sid/0’89 . .1 .7 7 734 JO 11 IS/VJU 7V/v/j DE N5 / TY~ MOISTURE RELA T/OIV PERCENT MOISTURE CONTENT * W57 8A5/5 FIGUR E / 7. SR and the expansion after release of the pressure. The density when pressure is applied is limited by the moisture content of the hay. when the hay is above 35 per- cent moisture content, it reacts as a hydraulic material as pressure is applied and will ooze from the cylinder through the clearances allowed for air. As pressure is increased- the hay will begin to spurt out and it then becomes impos- sible to hold the pressure. As moisture content is increased, the maximum pressure obtainable without loss of material from the cylinder is de- creased. This has made it difficult to obtain accurate re— sults with high moisture content hay. when material is forced from the cylinder during the pelleting process. the result is a poor quality pellet. Al- though one end is usually satisfactory, the other end is loose due to the lower pressure after the loss of material. When pelleting reconditioned hay the percent linear expansion is less for lower moisture content hay (Figure 16). This is due to the increased elasticity and the re- duced brittleness of the hay stems at higher moisture con- tents. This results in a higher density with pellets made from hey at higher moisture contents: The density-moisture rela- tionship for reconditioned hay (Figure 17) indicates that a decrease in moisture content from 30 to 20 percent before 55 pelleting would result in an increase in density from 12.6 to 22.1 pounds per cubic foot. These values were found by multiplying the values found on Figure 17 for density by a conversion factor of 3.813 to obtain density in the more common units of pounds per cubic foot. When reconditioned hay is above 30 percent moisture content, a quantity of liquid is squeezed from each pellet as it is made. The amount of liquid increases as the pres- sure and moisture content are increased. Although this in- volves a loss of nutrients, this loss is not important since the hay is dry when taken from storage and need only be brought up to 15 to 18 percent moisture content for pellet- 1n8.. The time required to apply pressure to the pellet to obtain a dense, firm pellet indicates a need for an addi- tional binding material when pelleting reconditioned hay. The expansion increased when pressure was held for less than two minutes and no firm pellets were obtained when pressure was held for less than one minute. This indicates that the efficiency of pelleting reconditioned hay would be increased with the addition of a binding material. The need for an additional binding material is due to the loss while in storage and during the reconditioning pro- cess of sugars and starches (12) necessary for creating a bond between the hay stems. 56 This would also be a method of adding to the feeding value of old hay. since Grahek (8) suggested several high energy ingredients for use as binders. Hidds, millrun, shorts and other forms of wheat starch were listed as desir- able binding materials. Regults of Pelletigg Gregn Bay. The linear expansions as well as the final densities were computed from the length and diameter measurements of the pellets. The percent mois- ture content of the pellets at the time of pelleting was computed from the wet and dry weights. A graph was made of the final density versus the mois- ture content of the pellets. In order to represent the density-moisture relationship, curvilinear regression lines were computed for the 3000. 5000 and 8000 pound per square inch pelleting pressures. A parabola was used to obtain the following equations; at 3000, x = 9.097 - 0.3u12x + 0.00h5uux2’ at 5000, - Y = 10. 896 - 0. #2362: 4 0.005613x2, and at 8000, I = H.053 - 0.6u7ux + 0.01038 1:2 , Where Y - densitngrams per cubic inch X - moisture content-percent 57 These results indicated that an exponential equation might be a more accurate representation of the density- moisture relationship. especially at higher moisture contents. The parabolas can not be used to predict density at higher moisture contents because of the minimum points. The following equations resulted when an exponential equation was used to represent the density-moisture rela- tionship and these are shown on Figure 19. At 3000 I = e 2.0305 - 0.0276 X, at 5000 D and at 8000 Y . e 2. 568 - 0.01405“. These equations indicate a definite linear relationship between the pressure used in pelleting and the computed co- efficients. The parabolas indicated a relationship. but it was not definitely linear. The least squares method was used to get an estimate of the effect of pressure on the coefficients of the equation I=°a+bx. (1) fallowing Z to represent pressure in pounds per sqaure _inch,a=AZ+Bandb=CZ+D. (2.3) Substituting these into (1) results in I=e(AZ+B)+(CZ+D)X. (1+) DENSITY- MOISTURE RELATION GREEN HAY 32§e§§ ' l t. §§g§es / N '9 § to ecln / l l /?45 l l : + I/ l I I d <1 x o c + ‘o /' .l (I I/ E g z [I °4 8 :3) [8 a x/ // c g 3 / if a . Q. /, // 0 § § at / II“ 9, 2“ ,3 x’ c)/ V Cb Q d7 j/ a m 7 ll (Q) I] 7/d u ' // // x >~ >~ x // W /// <4 a/I/ , 7 . “i W. I4 . /+ 0/ d/a O/ no): / n // u 7/ - / + // / /o / *2" ° (0 I // X /f [/ A? h/ flgi/ / / / I f/ /G l : 2 /oz Q: R m '0 v- 0: w Halv/ 3/9/79 83d SWVUQ 1.77734! .10 xuszvsa 7VIVIJ g 20 25' 30 35 PERCENT MOISTURE CONTENT [5 IO - WET BASIS FIGURE /9, 59 Solving (h) gives .AD D A x - e (3 ~ 0') + c(z + E)(x + ). The equation for the pellet density-moisture-pressure relationship with the computed coefficients is x 3 e 0.805 - 0.0000026s(z + 7360)(x.. #2.?) Although.this is an estimate of the real relationship, a more rigorous representation would be given by computing a multivariate regression with a greater number of samples using the equation 1°81=B+AZ+CZX+DX involving four variables. A variation of time of pressure application or maturity and quality of alfalfa would alter the coefficients in any of these equations. A graph was made of the moisture content versus the percent linear expansion (Figure 18). Regression lines were computed for the 3000. 5000 and 8000 pound per square inch pressures with an exponential equation. These equations showing the pellet expansion-moisture relationship were; at 3000 x 3 e 1.20 + 0.00805! J at 5000 x.= 9 1.151 + 0.010u61 ) 60 LIA/EAR fXPANS/O/V-MQQI’UHE HELIX 770/V GREEN .mdhefi. ants .. #29360 “squandered .9 mmbmt szUQWQ or on , o... v S , o _ M . L e. a _ _ . Random. 1. III: x mxccfmx c + \WQQOQW _u \ ..V Ransom ollkmvoSdtgl m u R .\\e\ x. \ e \MKQQQV x . x x c\i\\N\ . \WQQQQD all \h hQV§§§$QN